Thrombosis means the production of a blood clot, which is an endpoint component created when blood is styptic during blood coagulation processes, in the heart or in the blood vessels [1]. Normally, the dissolving process naturally eliminates blood clots, but it cannot dissolve all of the blood clots formed in large quantities by abnormal health conditions or chronic diseases. These remaining blood clots block the blood vessels and cause various dangerous diseases such as arteriosclerosis [2, 3]. For this reason, various modalities such as blood circulation improvement drugs are actively developed to remove blood clots accumulated in the blood vessels [4-6], and advanced medical techniques such as catheter interventions [7], stent installations [8] and real-time magnetic resonance or x-ray blood vessel imaging [9, 10] are widely investigated for diagnoses and treatment of diseases caused by remaining blood clots.

On the other hand, artificial thrombosis, which means a technique of creating blood clots artificially in a particular area, has been also developed in several preclinical animal models to identify and to analyze thrombotic problems, such as arteriosclerosis and stroke, or to verify methods for diagnosis and treatment of illnesses from the accumulated blood clots (or ‘the blood clots accumulation’) [11-14]. Photothrombosis is one of the methods to artificially generate blood clots in blood vessels in a specific region using photochemical activations. When a green LASER light illuminates the region of interest combined with Rose Bengal injection into the venous system, artificial thrombosis occurs in the target region [15-19]. The Rose Bengal is a chemical agent that is activated a green LASER. Photothrombosis has advantages in comparison with other methods in artificial blood clot generation: no direct, mechanical contact with blood vessels and tissues and a possibility in precise controls of locations of thrombosis occurrence. For these reasons, photothrombosis has been widely employed to develop animal models with blood vessel blockages inducing illnesses, for instance, an ischemic stroke [20-23].

Methods to verify the ischemic stroke animal model with thrombosis-based brain damage are also important. These verification methods can be mainly divided into two categories: histological assays and transient imaging. In a histological confirmation of ischemic stroke animal models, Triphenyltetrazolium chloride (TTC) staining is generally applied to observe the apparent color differences in the brain damage and to measure its size [24, 25]. The ex-vivo histological method requires sacrificing animal models for verifications of brain damages, leading to the inability to analyze transient brain damage development in the same animal model or to verify time-variant changes of brain damage with novel treatments. One of the methods to accurately measure changes of brain damage over time, with a minimal invasiveness, is magnetic resonance imaging (MRI) [26-30]. However, since it is hard to acquire MR (magnetic resonance) images at the same locations from the model animal during the thrombosis progress, it is difficult to see the transient effect of artificially induced thrombosis on the conventional MRI system. In previous studies of photothrombosis inducing and preclinical assays using the model animal with brain damages by photothrombosis, methods and tools for precise positioning of magnetic resonance imaging and obtaining as similar slices as possible are not actively employed [31-33]. A few studies of improvements in optical systems to ensure stable location or irradiation power of photothrombosis inductions [34, 35], but the systemic improvements on this issue have not been developed. Therefore, it was necessary to improve the optical system for accurate measurements.

To overcome this shortcoming, we investigated a photothrombosis inducing system integrated with nonmagnetic stereotaxic equipment for spatiotemporally aligned measurements of brain damage in ischemic stroke model animals. Precise positioning using the MR-compatible stereotaxic frame in the photothrombosis inducing system can offer a novel approach to acquiring slices with less variation in transient imaging of living model animals. We observed that the brain damage was well expressed in the model animals on the MRI system with the new apparatus. Later, we verified the result with assays using TTC. The system precisely formed brain damage in specific areas in rats, which was transiently verified by MRI. In addition, occurrence of brain damage and its size were validated by the TTC-based histological assay. We strongly believe that the system is capable of acquiring time-variant and well-aligned structural and functional imaging on various animal models established by optical stimulation. Also, we expected that transient MRI with more precise positioning is obtained when the nonmagnetic stereotaxic frame can be mounted with a customized MRI coil.

The photothrombosis inducing system for precise positioned MRI is illustrated in Fig. 1(a). The system consists of two parts: (1) a nonmagnetic stereotaxic frame to maintain stable positions in both photothrombosis and transient MRI and (2) an optical irradiation platform to induce photothrombosis on a specific region of the brain in rats. We applied the nonmagnetic stereotaxic frame (68011, RWD Life Science, Shenzhen, China) with fixation compartments that hold the rat by its mouse and ears and a customized acrylic baseplate (Hayoung Infra., Daegu, Republic of Korea) to minimize interfering noise between electromagnetic fields and the frame. In the optical platform, we employed a fiber-coupled LASER diode module (GL532T3-200FC, Shanghai Laser & Optics Century Co., Ltd., Shanghai, China) with 532 nm peak wavelength and output power controls. Light from the LASER source was delivered through a FC/PC optical fiber (MM-200µm-DPC, Shanghai Laser & Optics Century Co., Ltd.) with 200 µm core diameters and a fixed focus collimator with a FC/PC connector and f = 7.86 mm (F240FC-532, Thorlabs, Newton, New Jersey, United States). After that, light was focused on the selected region in the rat brain by a plano-convex lens with f = 35 mm (LA1027-ML, Thorlabs).

[FIG. 1.] (a) A schematic of a photothrombosis inducing system for precisely positioned magnetic resonance imaging. (b) A diagram of how a stereotaxic frame holding a rat moves to a knee coil in a magnetic resonance imaging system after a generation of brain damage by photothrombosis.

To adjust precise positions of light irradiation to generate localized photothrombosis, a two-dimensional translation manual stage (SU4-100, Science Town, Incheon, Republic of Korea) with 25-mm micrometer heads was applied for lateral (XY) positioning. And a motorized stage (MTS50/M-Z8, Thorlabs) with an electrical controller (KDC101, Thorlabs) was employed for precise light height (Z, focus) adjustments. To connect each compartment in optical irradiation platform, posts and other optical components (Nam-il Optical Instruments Co., Incheon, Republic of Korea) were used. Also, a C-clamp (SM-FTU80, SMATO, Seoul, Republic of Korea) was used for positioning two parts. The output optical power produced in the system with controls of a knob in the LASER diode module was measured by a photo-detector (S305C, Thorlabs) connected to an USB-linked optical power meter (PM100USB, Thorlabs). We conducted a brain damage generation experiment with various laser irradiation optical powers to determine the condition that light could be transmitted through a skull and thus confirm the output that could cause brain damage in a precise, local area. We checked that irradiation optical power was 17 mW for photothrombosis induction.

Four male, 7-12 weeks Sprague Dawley (SD) rats, which were employed for a feasibility study of the photothrombosis inducing system for precise positioned MRI, were acquired from Koatech, Pyeongtaek, Republic of Korea. The characteristics of brain damage inducement in animals were derived from previous studies of preclinical brain damage model investigations and applications [36-40]. This research was approved by the Institutional Animal Care and Use Committee of Daegu-Gyeongbuk Medical Innovation Foundation (Approval number: DGMIF-17073101-01). Rats were anesthetized by an intraperitoneal (IP) injection of 30 mg/kg Zoletil (Zoletil 50 inj., Virbac Korea, Seoul, Republic of Korea) and 10 mg/mL Rompun (Rompun inj., Bayer Korea, Seoul, Republic of Korea) before the induction of photothrombosis. After the anesthesia, we trimmed the hair on the skin of the skull and cut the center part out with a disinfectant using Povidone-iodine swabs (A17200231, Dongindang, Seoul, Republic of Korea). Rats were fixed to the non-magnetic stereotaxic frame in the photothrombosis inducing system, a brain periosteum was cut apart, and the skull was exposed.

The probe to irradiate LASER for photothrombosis was laterally aligned by controlling manual translation stages for producing brain damage in the area associated with the exercise in the hind legs. The height of the probe was also axially adjusted by the motorized stage in the photothrombosis inducing system for maximizing the transfer efficiency of the light. After the procedures to align the probe, the Rose Bengal (330000, Sigma-Aldrich, St. Louis, MO, United States), which was diluted with saline to 10 mg/mL, was injected through a tail vein with a dose rate of 80 mg/kg for 2 minutes while the other light was shaded and the LASER light to generate photothrombosis was exposed. A total light exposure time to generate photothrombosis was 20 minutes and the skull was sealed by a stitching fiber after the light exposure.

After induction of brain damage by the photothrombosis system was finished, the stereotaxic frame and the rat were positioned in a specific region of the coil integrated to the MRI system. In this study, we employed the 3-tesla MRI system (MAGNETOM Skyra, Siemens, Erlangen, Germany) and the transmit/receive 15-channel knee coil (Siemens) with 15 integrated preamplifiers, elements arranged in 3 rings by 5 elements. The MRI was conducted immediately (0 hours), 24 hours, 48 hours, and 72 hours after the induction of photothrombosis.

After the induction of photothrombosis, the MR-compatible stereotaxic compartment with the rat was removed from the photothrombosis induction system and moved into the coil of MRI. For a tight alignment of the position, an ear fixture in the stereotaxic compartment, a marker in the knee coil, and cross-shaped laser guide in the MRI were laterally matched. The horizontal position is checked by a level meter that can be used in the MRI. To confirm a location of the center and a region in imaging, a high-speed MR image acquisition was performed prior to each imaging, followed by T1- and T2-weighted imaging. While this confirmation should be done each time when the images are acquired in a general case where the system was not employed, the photothrombosis inducing system allowed the acquisition of a highly consistent slice through parameters obtained by the pilot image at 0 hours. A Java-based image processing program (ImageJ^TM) was employed as a tool for image verifications and analyses.

To confirm brain damage induced by the photothrombosis inducing system, the TTC (T8877-25G, Sigma-Aldrich) staining method was employed. After three days of MRI brain damage monitoring, the brain was extracted from the rat and sliced every three millimeters using a coronal rodent brain matrix. And for each slice, it was stained by TTC to visualize the brain damage. Based on the sufficient correlation between T2-weighted MR images and histological assays, derived from previous preclinical studies [41-45], staining verification took as an end-point measurement after MR imaging at 72 hours. The overall and detailed steps in generation of brain damages by photothrombosis, transient and non-invasive MR imaging, and histological identification of brain damages by TTC staining were illustrated in Fig. 2.

[FIG. 2.] A schematic of procedures in generation of brain damage by photothrombosis, transient and non-invasive MR imaging, and histological identification of brain damages by TTC staining.

In MR imaging, T1-weighted images can provide structural changes or anatomical information and T2-weighted images offer more reliable information on lesions [46, 47]. To be detailed, T1-weighted images are acquired by short time to echo (TE) and short repetition time (TR) for minimizing T2 relaxation [48, 49]. On the other hand, T2-weighted images are captured using longer TE and TR for minimizing T1 relaxation effects. According to these settings, tissues with a high proportion of fat are bright in T1-weighted images and locations with a high proportion of water components (for instance, lesions) are bright in T2-weighted images. Figure 3 indicates transient T1- and T2-weighted images of the brain in the rat after the photothrombosis. T2-weighted images of the rat brain indicate that brain damage occurred strongly 24 hours after the optical thrombosis by the photothrombosis inducing system, because it takes time for brain damage to be generated due to blocked blood vessels in LASER light illuminated areas. Compared to the higher changes of intensities in T2-weighted images, the change in T1-weighted images is insignificant. It means that structural and anatomical damage in the rat brain is little when photothrombosis was introduced. Also, there was a slight difference in size, but the brain damage was maintained after 72 hours of LASER light irradiation. It indicates that the animal model with brain damage, which was generated by the photothrombosis inducing system, is available to compare and verify the effectiveness of the various treatment modalities.

For a quantitative comparison in intensities of MR images, manually segmented measurements of average intensities for T1- and T2-weighted images for photothrombosis induced, non-induced (the other side of the rat brain) and background regions are shown in Fig. 3. A size of the segmentation is 1.56 × 1.56 mm² and central points (as illustrated in Fig. 4(a)) of photothrombosis induced, noninduced and background regions are (x = 47.88 mm, y = 45.42 mm), (x = 53.90, y = 45.87) and (x = 99.22 mm, y = 0.78 mm) when the top left is set to (x = 0.00 mm, y = 0.00 mm). Figure 4 represents average intensities of T2- (Fig. 4(b)) and T1- (Fig. 4(c)) weighted magnetic resonance images on the photothrombosis induced, non-induced, and background regions. An error bars in each bar represents the standard deviation for the intensity in the measurement area established in Fig. 4(a). Comparing the significant changes (2.108 times) in T2-weighted images between 0 and 24 hours after the photothrombosis induction and the minute changes in T1-weighted images, the photothrombosis inducing system can produce an ischemic brain damage in the specific region without structural brain changes. When compared to T2 images of the brain tissue in the region without the photothrombosis induction, the occurrence of increasing average intensities due to the brain damage by the photothrombosis was verified. Also, in analyzing MR images presented in Fig. 3 and data in Fig. 4, it was determined that the validity analysis of the treatment device of brain damage was most appropriate for photothrombosis induced rats after 24 hours.

[FIG. 3.] Transient T1- and T2-weighted magnetic resonance images of the brain in the rat acquired immediately (0), 24, 48, and 72 hours after the photothrombosis.

[FIG. 4.] (a) Estimated of regions of manually segmented measurements of average intensities in magnetic resonance images (①: a photothrombosis induced region, ②: a non-induced region of the rat brain, ③: a background region). Average intensities of (b) T2- and (c) T1-weighted magnetic resonance images on photothrombosis induced (black), non-induced (red), and background (blue) regions. Each average intensity was determined by manually segmented measurements with a size of 1.56 × 1.56 mm2. An error bar in each bar represents the standard deviation for the intensity in the measurement area established.

Figure 5 represents three-dimensional projected images of the rat brain after 0 (as shown in Fig. 5(a)) and 24 hours (as illustrated in Fig. 5(b)) of brain damage generation by the photothrombosis inducing system. A three-dimensional projected images show that brain damage had been well induced around the area where the LASER light was in contact. It is expected that a three-dimensional image projection and region analysis can allow verification of various treatment methods of brain damage caused by cerebral diseases.

[FIG. 5.] Three-dimensional projected images of the rat brain after (a) 0 and (b) 24 hours of brain damage generation by the photothrombosis inducing system.

Figure 6 represent an overall photograph of the rat brain that was removed 72 hours after brain damage generation by photothrombosis and a result of histological confirmation by TTC staining and brain slicing. The overall photograph of the rat brain (as illustrated in a left figure in Fig. 5) indicates that the brain damage was precisely generated in the specific region of a LASER irradiation. The result of TTC staining and brain slicing for confirming brain damage generations (as described in a right figure in Fig. 6) also represents that the left front part of the rat brain (marked with red arrows) is white. Since TTC staining means the release of dehydrogenase of the mitochondria in tissues, the unstained region is the area of tissue injuries by photothrombosis. Although it is difficult to compare the exact size due to differences in the exact position and thickness of the slices in MRI and histological assays, it was possible to cross-check that brain damage occurred in the aiming position using photothrombosis.

[FIG. 6.] An overall photograph of the rat brain that was removed 72 hours after brain damage generation by photothrombosis and a result of histological confirmation by TTC staining and brain slicing.

After the photothrombosis inducing system was investigated, the animal models with brain damage were developed, and the system was improved and supplemented based on the process and results in photothrombosis. In order to develop more precise and quantitative animal models with brain damage, we strongly believe that there should be several points to be improved in the photothrombosis inducing system. In the positioning of LASER light irradiation in the experiment, a stage equipped with a motor and a controller was employed to align the distance between the brain of the rat and the light source as precisely as possible. For the development of a more precise positioning, distance information shall be acquired by sensors for distance measurements. Also, when inserting the stereotaxic frame with the rat into the coil and adjusting its position, there is some difference in imaging position, although the adjustment is made as precise as possible using laser guides in the MRI instrument and several markings. Although the difference is smaller than without the stereotaxic frame, we expect to further reduce the difference in imaging regions and locations of the center when MRI coils to mount the stereotaxic frame correctly are employed. As an example, the method of applying birdcage coils with multiple ports to the photothrombosis inducing system can be useful in that they can simultaneously measure multiple brain damage animal models under different conditions [50]. It is also expected that developing the system that combines magnetic resonance imaging and optical hemodynamic imaging, for instance, optical coherence Doppler imaging [51], photoacoustic functional imaging [52], and diffusion optical brain tomography [53], can be applied in quantification and optimization of animal models with ischemic brain damage. Also, optical imaging with constructions of thin and precise brain slices and label-free deep tissue structural imaging techniques can be applied to an accurate and detailed analysis of histological assays [54-56].

When photothrombosis is applied to animals that are not small animals, it is generally necessary to remove a part of the skull to expose brain tissue and illuminate with light. For instance, Ikeda et al. introduced a non-human primate model with a brain damage by photothrombosis with a 532 nm laser applied to the exposed area of the skull [57]. For minimizing the discomfort and potential risks of the skull removal, we propose several techniques that can penetrate or focus light through the scattering layers can be employed as a noninvasive method to induce photothrombosis for animals with thick skulls [58-63].

In this study, we investigated a photothrombosis inducing system integrated with nonmagnetic stereotaxic equipment for spatiotemporally aligned measurements of brain damage in ischemic stroke animal models. Precise positioning using a frame in the system provides an advanced approached to obtaining image slices with less variation in transient measurements of brain damage. T1- and T2-weighted MRI can provide transient information in incidence, location, and size of brain damage generated by photothrombosis using the system. Also, congruent ischemic brain damage was confirmed by the TTC staining-based histological assay. The system is useful in that it provides the more reliable level of reproducibility in a spatial view of MR imaging, while enabling brain damage observations over time. More precise, quantitative, and optimized investigations of animal models with brain damage derived by photothrombosis require improvements of the photothrombosis inducing system and coupling with MR coils. We expected that convergences of optical hemodynamic imaging modalities with MR imaging can offer more detailed and complex information to be used for optimized animal model developments.